INTRODUCTION

Top Abstract Introduction Materials and Methods Results Discussion References

Fusarium subglutinans f. sp. pini is an important pathogen of pine that causes pitch canker in mature trees (6, 13) and root rot and damping-off in seedlings (2, 34). This fungus can be spread by both infected seedlings and seed (1, 28). The management of F. subglutinans f. sp. pini would be greatly improved if a quick screening method were available for seed and nursery stock.

F. subglutinans f. sp. pini represents one of eight mating populations (i.e., biological species) in the Gibberella fujikuroi species complex (6, 23). Three of these mating populations, B, E, and H (F. subglutinans f. sp. pini), have F. subglutinans anamorphs (5, 14, 19, 20). Strains of Fusarium isolated from pineapple (F. subglutinans f. sp. ananas) and mango, for which a teleomorph is not known, also have F. subglutinans anamorphs (27, 32, 33).

Distinguishing F. subglutinans f. sp. pini from the other species in Fusarium section Liseola usually requires pathogenicity tests or sexual crosses with known tester strains (6, 7, 35). These assays are time-consuming and labor-intensive and do not always yield clear-cut answers. Molecular tools such as random amplification of polymorphic DNA (RAPD) (9, 35, 36), mitochondrial restriction fragment length polymorphisms (RFLP) (7), and ribosomal DNA (rDNA) internal transcribed spacer (ITS1 and ITS2) sequences (25, 37) have been tested for their efficacy in differentiating F. subglutinans f. sp. pini isolates from other isolates of F. subglutinans. Because of the technical difficulties associated with mitochondrial RFLP and the low repeatability of RAPD data, we do not consider these techniques useful for diagnostic purposes. Two different copies of the ITS2 region were identified in the same isolate within some of the species in Fusarium section Liseola (25, 37), and a reliable diagnostic technique based on these sequences could not be developed. Alternative regions such as the histone and -tubulin genes might be used more effectively.

O'Donnell et al. (26) used the DNA sequences of the nuclear rDNA large subunit, mitochondrial small subunit, and -tubulin to develop a phylogeny that includes 36 taxa in the G. fujikuroi species complex. These sequences may potentially be useful for diagnostic purposes, but we began our study prior to publication of the phylogeny of O'Donnell et al. (26).

We used an alternative region of the genome, the histone H3 gene, to distinguish F. subglutinans f. sp. pini isolates from other isolates of F. subglutinans. Histone genes encode histone proteins, which are the major constituents of chromatin (16, 21). Four histone proteins, H2A, H2B, H3, and H4, make up the nucleosomal core (17). The gene encoding the H3 protein is well conserved, especially at the amino acid level (12, 31), and the presence of introns enhances its value in taxonomic and phylogenetic studies of closely related organisms (8, 38). Although the histone H4 gene also has these characteristics, it is generally too highly conserved to be suitable for evolutionary studies (30).

Our objectives in this study were (i) to sequence the histone H3 gene from various strains in the G. fujikuroi species complex, (ii) to compare the relationships thus determined with those established by use of other sequences, and (iii) to develop a PCR-RFLP procedure based on the histone H3 gene sequence for the routine identification of F. subglutinans f. sp. pini.

	MATERIALS AND METHODS

Top Abstract Introduction Materials and Methods Results Discussion References

Fungal isolates. All isolates were maintained on 2% (wt/vol) malt extract agar (Biolab Diagnostics Ltd., Fedlife Park, Midrand, South Africa) in the culture collections of the Forestry and Agricultural Biotechnology Institute at the University of Pretoria, Pretoria, South Africa, and the Medical Research Council, Tygerberg, South Africa. We examined 42 Fusarium isolates, including F. subglutinans f. sp. pini, pathogenic to pine; F. subglutinans f. sp. ananas, pathogenic to pineapple; F. subglutinans isolates associated with maize and mango; and the mating type tester strains from all eight mating populations in the G. fujikuroi species complex (Table 1). To test the efficacy of the PCR-RFLP technique for use as a species diagnostic technique (see below), we tested 60 strains of the H mating population identified by Britz et al. (5) and 80 strains representing populations A to F identified by Yan et al. (39). These strains were reassorted and then encoded so that the assays were done in a blind manner.

DNA isolation. Flasks containing 100 ml of malt extract broth (2% [wt/vol]) (Biolab) were inoculated with 1-ml spore suspensions (>1,000 spores/ml). After 2 weeks of static incubation at room temperature (20 to 25°C), mycelium was harvested by filtration through no. 1 filter paper (Whatman BioSystems Ltd., Maidstone, Kent, United Kingdom). Harvested fungal tissue was ground to a powder in liquid nitrogen with a mortar and pestle and homogenized in extraction buffer containing 5% (wt/vol) CTAB (N-cetyl-N,N,N-trimethylammonium bromide), 1.4 M NaCl, 0.2% (vol/vol) 2-mercaptoethanol, 20 mM EDTA, 100 mM Tris-HCl (pH 8.0), and 1% (wt/vol) polyvinylpyrrolidone. This homogenate was incubated at 60°C for 1 h and centrifuged (16,000 × g) at room temperature. We performed phenol-isoamyl alcohol-chloroform (25:1:24) extractions and removed residual phenol with an additional chloroform extraction. Nucleic acids were precipitated by adding 0.1 volume of 3 M sodium acetate (pH 5.2) and 0.6 volume of 2-propanol, followed by incubation at 4°C overnight. Precipitated DNA was centrifuged (16,000 × g), washed with 70% ethanol, and resuspended in deionized water. This protocol is a variation of the one developed by Murray and Thompson (22).

PCR amplification. PCR amplification was performed as described by Glass and Donaldson (12) with primers H3-1a (5'-ACTAAGCAGACCGCCCGCAGG-3') and H3-1b (5'-GCGGGCGAGCTGGATGTCCTT-3'). These primers were constructed to flank at least one intron and amplify approximately 450 bp of the Neurospora crassa histone H3 gene. Each PCR mixture contained 1 mM deoxynucleotide triphosphates (0.25 mM each), 2.5 mM MgCl₂, 0.2 µM H3-1a, 0.2 µM H3-1b, 0.25 ng of DNA per µl, 0.05 U of Super-Therm DNA polymerase [Southern Cross Biotechnology (Pty.) Ltd., Cape Town, South Africa] per µl, and 1× Super-Therm reaction buffer. PCR mixtures were overlaid with mineral oil, and reactions were performed on an Omnigene thermocycler (Hybaid, Middlesex, United Kingdom) with an initial denaturation step of 1 min at 92°C. This step was followed by 30 cycles of denaturation at 92°C (1 min), annealing at 68°C (1 min), and elongation at 72°C (1 min). A final extension was performed at 72°C for 5 min.

DNA sequencing. PCR products were purified with a QIAquick PCR Purification Kit (Qiagen GmbH, Hilden, Germany). Histone H3 gene fragments from the 42 Fusarium isolates included in this study, were sequenced (see Table 1 for GenBank accession numbers) in both directions with primers H3-1a and H3-1b. Reactions were performed on an ABI PRISM 377 automated DNA sequencer with an ABI PRISM Dye Terminator Cycle Sequencing Ready Reaction Kit (Perkin-Elmer, Warrington, United Kingdom).

Sexual compatibility tests. The seven F. subglutinans isolates recovered from maize in South Africa (Table 1) were crossed with mating population E tester strains and with one another in all possible pairwise combinations (5, 18). Crosses were scored as positive when ascospores were observed exuding from perithecia.

PCR-RFLP technique. Amplified DNA was digested with two restriction enzymes, CfoI and DdeI (Boehringer Mannheim South Africa Pty. Ltd.). Digestions were performed consecutively by adding 5 U of CfoI to 15 µl of unpurified PCR product (3); after 3 h of incubation at 37°C, 5 U of DdeI was added and the sodium chloride concentration was adjusted to 100 mM. These digestion reaction mixtures were then incubated at 37°C for an additional 5 h. We resolved PCR-RFLP profiles on 3% (wt/vol) agarose gels (Promega Corporation, Madison, Wis.; molecular biology-grade agarose) containing ethidium bromide (0.2 µg/ml). Electrophoresis was performed at 3 V/cm (room temperature) with 0.5× electrophoresis buffer containing 4.5 mM Tris, 4.5 mM boric acid, and 1 mM EDTA (pH 8.0). Nucleic acids were visualized with a UV transilluminator (302 nm).

Verification of technique. To test the efficacy of the PCR-RFLP technique described here, histone H3 gene PCR products from 60 strains representing mating population H and 80 strains representing mating populations A to F were amplified, digested, and electrophoresed as described above. We compared the resulting PCR-RFLP profiles to those generated from the representatives of the G. fujikuroi species complex.

	RESULTS

Top Abstract Introduction Materials and Methods Results Discussion References

DNA sequencing. The Fusarium histone H3 gene fragments ranged from 519 to 527 bp in length and contained two introns (intron 1 and intron 2) whose positions within the sequences were conserved. Intron 1 was 83 bp long for strains from mating population H, F. oxysporum, and F. subglutinans f. sp. ananas; 81 bp long for mating populations C and D and F. subglutinans isolated from mango; 85 bp long for mating populations A and G; 82 bp long for mating populations E and F and F. subglutinans isolated from maize; and 77 bp long for mating population B. Intron 2 was 57 bp long for all of the isolates, except for F. oxysporum, for which it was 58 bp long.

View larger version (46K): [in this window] [in a new window]   FIG. 1.   Phylogram generated with histone H3 gene sequence data from the isolates included in this study by use of PAUP 4.0b1. Bootstrap values based on 1,000 replications are indicated as percentages. Bold letters in parentheses refer to the G. fujikuroi mating populations. This dendrogram is rooted to F. oxysporum MRC 6212. The length of the tree was 201 steps, and the values for homoplasy index and the retention index were 0.24 and 0.94, respectively.

Sexual compatibility tests. Three of the F. subglutinans isolates associated with maize (MRC 1077, MRC 837, and MRC 714) were sexually compatible with one of the mating type tester strains for mating population E (MRC 6483). The remaining four isolates did not cross with one another or either of the tester strains.

PCR-RFLP technique. PCR-RFLP analysis of the amplified histone H3 gene products with DdeI and CfoI enabled us to distinguish F. subglutinans f. sp. pini from the rest of the isolates included in this study (Fig. 2). Unique PCR-RFLP profiles were generated for each group included in this study, except for mating populations C and D, mating population G, and F. subglutinans isolated from mango. From the restriction enzyme profiles, we constructed restriction maps for all host-specific groups of F. subglutinans as well as F. moniliforme, F. proliferatum, F. thapsinum, and F. nygamai (Fig. 3).

View larger version (60K): [in this window] [in a new window]   FIG. 2.   PCR-RFLP profiles generated by digestion of Fusarium histone H3 gene amplification products with DdeI and CfoI. Electrophoresis were performed on 3% agarose gels at 3 V/cm. Lanes M, 100-bp ladder (1,500, 1,000, 900, 800, 700, 600, 500, 400, 300, 200, and 100 bp); lane A, mating population A; lane B, mating population B (F. subglutinans associated with sugarcane); lane C, mating population C; lane D, mating population D; lane E, mating population E (F. subglutinans associated with maize); lane F, mating population F; lane G, mating population G; lane H, F. subglutinans f. sp. pini (mating population H); lane 1, F. subglutinans from maize; lane 2, F. subglutinans from pineapple (F. subglutinans f. sp. ananas); lane 3, F. subglutinans from mango.

View larger version (22K): [in this window] [in a new window]   FIG. 3.   Restriction maps of the histone H3 genes from the different isolates of Fusarium, generated with restriction enzymes DdeI and CfoI. The Fusarium introns are indicated as boxes, and the exons are indicated as horizontal lines. Bold letters refer to the G. fujikuroi mating populations. An arrow indicates a CfoI restriction site, and a vertical line indicates a DdeI restriction site. Exon and fragment sizes are indicated in base pairs.

Verification of technique. All 60 mating population H strains were positively identified as F. subglutinans f. sp. pini in a blind test of the PCR-RFLP technique. We identified none of the strains from the collection of Yan et al. (39) as F. subglutinans f. sp. pini, and the expected profiles were generated for each of their representatives of mating populations A, B, E, and F. The blind test of 140 samples was 100% successful, providing 95% confidence that the error rate for this test is less than 2%.

	DISCUSSION

Top Abstract Introduction Materials and Methods Results Discussion References

In this study, we were able to distinguish F. subglutinans f. sp. pini (mating population H) from F. subglutinans isolates associated with mango, maize (mating population E), sugarcane (mating population B), and pineapple and F. moniliforme (mating population A), F. proliferatum (mating populations C and D), F. thapsinum (mating population F) and F. nygamai (mating population G). The PCR-RFLP technique has been used successfully by the Tree Pathology Co-operative Programme diagnostic clinic to identify isolates of F. subglutinans f. sp. pini for the last year. Seven outbreaks of root rot in South African nurseries have been correctly diagnosed as being caused by F. subglutinans f. sp. pini (4). We thus have confidence that this technique is robust and can be used with a high degree of certainty.

Phylogenetic analyses with the Fusarium histone H3 gene sequence data generated a phylogram (Fig. 1) that was similar to those produced by O'Donnell et al. (26). The results presented here and those based on -tubulin and mitochondrial small-subunit DNA sequences (26) are similar to those obtained with isozymes (15) in two aspects. First, mating populations C and D form a closely related group in all cases. Second, mating population E is phylogenetically distinct from mating populations A, B, C, D, F, and G.

There are, however, two major differences between DNA-based phylogenies and those based on isozymes. With isozymes, Huss et al. (15) showed mating populations C and D to be most closely related to mating population G. The DNA-based phylogenies (26; this study), however, indicate that mating population G is most closely related to mating populations A and F and that these three mating populations form a distinct cluster separate from both mating populations C and D. Also, in contrast to the results from the isozyme study (15) both DNA-based phylogenies (26; this study) indicate that mating populations C and D are most closely related to mating population B.

F. subglutinans f. sp. pini has previously been reported to belong to mating population B (29), but our results and those presented by Britz et al. (5) and O'Donnell et al. (26) suggest otherwise. Nirenberg and O'Donnell (24) elevated this fungus to species level and provided the name F. circinatum (teleomorph = G. circinata) for it. Although our results are consistent with those of O'Donnell et al. (26) and support the placement of F. subglutinans f. sp. pini in a distinct taxon, the distinguishing morphological characters reported by Nirenberg and O'Donnell (24) appear to be inadequate to make definite identifications of the fungus (5).

F. subglutinans f. sp. pini, F. subglutinans f. sp. ananas, mating population E, and F. subglutinans isolated from maize are closely related to each other and are included in clade 1. Although some of the F. subglutinans isolates from maize and those belonging to mating population E appeared in two separate but closely related groups, this separation was caused by only two nucleotide base-pair differences. Since some individuals from both of these groups could cross with one of the mating type E tester strains, we do not believe that the second cluster of isolates from maize represents a separate mating population. The overall appearance of clade 1 corresponds to that of the so-called American clade described by O'Donnell et al. (26). This similarity suggests an equivalence of F. subglutinans f. sp. pini and F. circinatum as well as of F. subglutinans f. sp. ananas and F. guttiforme.

The two subgroups that constitute clade 2 in our study correspond to the African and Asian clades of O'Donnell et al. (26). The African clade includes mating populations A, F, and G, whereas the Asian clade includes mating populations B, C, and D. The latter clade also includes F. subglutinans isolates associated with mango, which are phylogenetically separate from F. subglutinans isolates associated with maize, pineapple, and pine but phylogenetically more closely related to F. subglutinans from mating population B (Fig. 1).

The results of this study and those of O'Donnell et al. (26) have identified a number of conserved genes that are useful for phylogenetic and taxonomic studies among species of Fusarium. The H3 gene, as well as the -tubulin gene, allows for a higher degree of resolution than rDNA ITS1 and ITS2. Species previously considered too closely related for separation into distinct groups can now be separated based on histone or -tubulin gene sequence. Moreover, rapid identification of fungi such as the pitch canker pathogen is now possible with a PCR-RFLP technique based on the histone H3 gene sequence.